Ikibazo gihuza ibibazo · Hagati Urwego

Ikurikiranyabihe-Kuri-Icyitegererezo Cyikibazo

Wige uburyo kodegisi-decoder yerekana ikarita yinjiza ikurikiranye kubisubizo birebire bitandukanye ukoresheje kwitondera no guhuza ibitekerezo.

Inzira zijyanye no kuyoboraUrukurikirane Kuri Moderi Ikurikirana
Ikibazo 1 Bya 8

Nibihe bintu bibiri byingenzi bigize urutonde rwikurikiranya?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Nibihe bintu bibiri byingenzi bigize urutonde rwikurikiranya?
    • Kodegisi na decoder
    • Ikimenyetso hamwe nicyiciro
    • Kode ebyiri
    • Akayunguruzo hamwe nu rutonde
  2. Ni ikihe kibazo cyingenzi uburyo bwo kwitondera bwakemuye muburyo bwa seq2seq?
    • Buhoro buhoro
    • Uburebure-burebure bwa vector icupa kubintu byinjira birebire
    • Kubura amagambo
    • Amakuru menshi yo guhugura
  3. Ni ubuhe bwubatsi bw'umwimerere wa seq2seq yo mu mwaka wa 2014 yakoresheje?
    • Imiyoboro ihamye
    • Ibiti
    • LSTMs
    • Bare ibiryo-byimbere
  4. Nigute kwitondera kubaka imiterere kuri buri ntambwe ya decoding?
    • Mugereranije ibyinjijwe byose bingana
    • Kwirengagiza kodegisi rwose
    • Mugukoporora ibimenyetso byanyuma
    • Numubare uremereye wa encoder leta ukoresheje amanota yo guhuza
  5. Kuki ibisubizo bya seq2seq bishobora gutandukana muburebure ninjiza?
    • Decoder itanga ibimenyetso kugeza bisohoye ikimenyetso cyo guhagarara
    • Ibyinjira nibisohoka burigihe buringaniye
    • Kodegisi yerekana ibintu byose
    • Uburebure bugenwa mbere
  6. Ni ubuhe bwoko bugezweho bukunzwe na Transformer ishingiye kuri seq2seq yubatswe?
    • ResNet na VGG
    • T5 na BART
    • K-bisobanura na PCA
    • Ijambo2Vec na GloVe
  7. In encoder-decoder attention, which representations are commonly used as keys and values?
    • Only the decoder’s future hidden states
    • The vocabulary file’s byte offsets
    • The encoder’s output representations
    • The optimizer’s learning-rate history
  8. What does a bidirectional encoder provide to an attention-based decoder?
    • A fixed one-token output regardless of the input
    • A guarantee that output length equals input length
    • A causal mask over future target tokens
    • Contextual representations of source tokens that can use information from both directions